基于神经网络预测方法的智能化高精度大型钢管高速托辊机构研究  

Research on Intelligent and High-Precision High-Speed Roller Mechanism of Large Steel Pipes Based on Prediction Method Using Neural Networks

作　　者：曲延涛[1] 杨风艳[1] 辛培刚[1] 王增波[1] 杨富广[1] QU Yantao;YANG Fengyan;XIN Peigang;WANG Zengbo;YANG Fuguang(Offshore Oil Engineering(Qingdao)Co.,Ltd.,Qingdao 266520,China)

机构地区：[1]海洋石油工程(青岛)有限公司,山东青岛266520

出　　处：《智能物联技术》2024年第4期25-31,共7页Technology of Io T& AI

摘　　要：大型钢管高速托辊机构在工作过程中受管节点偏心、滑动摩擦等不确定因素的影响,会使得主管转动距离与胶轮转动距离存在一定偏差。这种偏差给实际工程施工带来一定困难。对此,采用商用软件ANSYS中的Transient Structural模块建立基于某大型钢管高速托辊机构的有限元模型,通过仿真计算研究转动过程中主管和胶轮的运动特性,获得对应的旋转角度、旋转距离以及距离偏差,采用径向基神经网络方法对主管和胶轮旋转的偏差进行预测对比分析。研究结果表明,所建立的径向基神经网络方法能够很好地预测主管和胶轮旋转的转动距离偏差,能够为大型钢管高速托辊机构上主管与胶轮距离偏差的计算提供一定的参考。Due to the influence of uncertain factors such as tube joint eccentricity and sliding friction,the rotating distance of the main pipe will deviate from the rotating distance of the rubber wheel.This deviation brings some difficulties to the actual construction of the project.In this regard,the Transient Structural module in ANSYS was used to establish a finite element model based on a large steel pipe high-speed idler mechanism.The motion characteristics of the pipe and rubber wheel during rotation were studied through simulation calculation,and the corresponding rotation angle,rotation distance and distance deviation were obtained.The radial basis neural network method was used to predict and compare the rotation deviation between the main pipe and the rubber wheel.The results show that the radial basis neural network method can predict the deviation of the rotation distance between the main pipe and the rubber wheel well,and can provide a certain reference for the calculation of the distance deviation between the main pipe and the rubber wheel on the high-speed roller mechanism of large stel pipe.

关 键 词：大型钢管高速托辊机构 径向基神经网络 转动偏差 

分 类 号：TE928[石油与天然气工程—石油机械设备]